Group 1 Meredith, Barclay, Woo-je, and Kumar

1. Exploring Internal Stickiness: Impediments to the Transfer of Best Practices Within the Firm(Gabriel Szulanski, SMJ 1996) Group 1 Meredith, Barclay, Woo-je, and Kumar

2. Motivation • Impediments to the transfer of best practices BETWEEN firms has been examined, yet this has not been examined WITHIN firms • Transferring capabilities within a firm is not easy: • General Motors – difficulty transferring manufacturing practices between divisions (Kerwin & Woodruff, 1992) • IBM – limited success in transferring reengineered logistics and hardware design processes between business units (The Economist, 1993) • Thus, this paper provides us with an empirical investigation of internal stickiness

3. Some Definitions • Transfer of best practice inside the firm: To practitioners connotes firm’s replication of an internal practice that is performed in a superior way in some part of the organization and is deemed superior to internal alternate practices and known alternatives outside the company • Practice: refers to organization’s routine use of knowledge, often has a tacit component • Transfer of best practice (for this paper): dyadic exchanges of organizational knowledge between a source and recipient unit in which identity of recipient matters

4. Stages in the Transfer Process • Initiation: comprises all events that lead to decision to transfer (discovery of need, search for potential solutions, search leads to discovery of superior knowledge) • Implementation: begins with decision to proceed (resources flow between recipient and source, transfer specific social ties) • Ramp-up: begins when recipient starts using transferred knowledge (recipient concerned with identifying & resolving unexpected problems) • Integration: begins after recipient achieves satisfactory results (use of transferred knowledge gradually becomes routinized)

5. Analyzing difficulty of transferring practices within the firm • Notion of internal stickiness connotes difficulty of transferring knowledge within the organization • Some past researchers have referred to difficulty of transferring knowledge in terms of costs (Arrow, 1969; Teece, 1977; von Hippel, 1994) • Szulanski suggests cost could be a poor descriptor of the difficulty: • Deciding which portion of cost actually reflects difficulty is a matter of conjecture • Cost might fail to discriminate between problems that are equally costly but qualitatively different

6. Analyzing difficulty of transferring practices within the firm • Eventfullness: extent to which problematic situations experienced during a transfer are worthy of remark: • Could be translated into outcome-based descriptor of stickiness • Combined with stages, provides descriptor of stickiness for each stage • Initiation – problems stem from efforts to identify needs and knowledge that meets those needs • Implementation – communication problems or adaptation of practice to recipient’s needs • Ramp-up – struggle to achieve satisfactory performance • Integration – problems associated with achieving routinization

7. Origins of Internal Stickiness • Characteristics of knowledge transferred • Causal Ambiguity • Unprovenness • Characteristics of source of knowledge • Lack of motivation • Not perceived as reliable • Characteristics of recipient of knowledge • Lack of motivation • Lack of absorptive capacity • Lack of retentive capacity • Characteristics of the context • Barren organizational context • Arduous relationship

8. Empirical Method • 2-step questionnaire survey: • Self-selection of theoretically relevant companies (those with strong incentives to transfer best practices, were actively attempting to do so, and viewed it as important) • Model testing – 8 companies accepted for second step (AMP, AT&T, Paradyne, British Petroleum, Burmah Castrol, Chevron, EDS, Kaiser Permanente, Rank Xerox) • 271 returned questionnaires, 122 transfers of 38 best practices (both technical and administrative) • Companies instructed to rule out practices that could be performed by single individual and to choose only practices that required coordination of many people

10. Results • Explanatory power of the framework and relative importance of each barrier assessed by canonical correlation – can assess relationship between 2 SETS of variables • Explanatory Power • Canonical-R (0.87): Simple correlation between weighted sums of scores from each set of variables, using weights for first canonical root • Relative Importance of Each Barrier • Canonical weights reflect contribution of each construct to its canonical variate – weights can be compared and the larger the absolute value of a coefficient, the more important its contribution

11. Results Results suggest that 3 most important barriers are lack of absorptive capacity of the recipient (0.54), causal ambiguity (0.34), and an arduous relationship between source and recipient (0.33).

12. Limitations • Correlational design does not imply causation • Survival bias – problematic or aborted transfers not included. Only transfers that occurred were included.

13. Implications of Results • Practitioners and conventional wisdom attribute stickiness almost exclusively to motivational factors • Researchers agree with the dominant view of practitioners (e.g. Porter (1985) ) • Yet, Szulanki’s (1996) empirical results suggest that knowledge related barriers (lack of absorptive capacity, causal ambiguity, arduousness of relationship) dominate motivation-related barriers